The RMAA (RightMark Audio Analyzer) Source and Audio Device Measurement Thread [Overview of Measurements in Post #3, Tutorial in Post #2]

Discussion in 'Sound Science' started by hifichris, Mar 1, 2016.

  1. HiFiChris Contributor
    Noted with thanks and added to the overview.

    Seems like Sony still doesn't get the output impedance right with every device...
     
  2. HiFiChris Contributor
    Stoner Acoustics UD125:


    No load:

    no load.jpg


    All 4 Digital Filters (unloaded):

    all filters.jpg

    Default = minimum phase, sharp roll-off
    Filter 1 = minimum phase, slow roll-off
    Filter 2 = linear phase, slow roll-off
    Filter 3 = linear phase, sharp roll-off


    Loaded w/ Triple.Fi 10:

    triplefi10.jpg

    => Excellently low output impedance of ~ 0.1 Ohms.
     
    hakuzen likes this.
  3. HiFiChris Contributor
    HiFiMan MegaMini:


    No Load:


    No Load.jpg


    Triple.Fi 10:

    TF10.jpg
     
    Last edited: Jun 22, 2017
    hakuzen likes this.
  4. hakuzen
    JDS Labs C5D:

    DAC/Amp. Phones output. Output impedance: 0.62Ω @1kHz.

    No load:

    Normal (no bass boost):

    C5D.png

    Medium and high Bass boost:

    C5D-bass1-bass2.png


    Pai Audio MR3:

    C5D-MR3.png


    Amplifying E-MU 0404 (no load):

    EMU_vs_EMU-C5D.png


    Pai Audio MR3:

    EMU-C5D-MR3.png


    RMAA:


    ADC used: Creative E-MU 0404 (Z=1MΩ).
    Measured C5D headphones amplifier (phones output).
    Sources:
    - C5D analog input (aux), amplifying E-MU 0404 DAC (line out) .
    Drivers used: E-MU's ASIO.
    Output voltages: 1Vrms, 1.5Vrms, 2Vrms, 3Vrms.
    - C5D DAC (usb).
    Drivers used: ASIO4ALL.
    Output voltages: 1.060Vrms, 1.495Vrms, 2.108Vrms, 2.962Vrms.

    C5D-aux+usb_rmaa.png



    NOTES:
    Small spikes in noise (high frequencies) come from E-MU 0404's first input channel when using 48kHz and 24bit.
    Some distortion (3rd and last harmonics) come from E-MU 0404's input channels.


    Noise level:

    C5D-aux__noise.png

    Dynamic range:

    C5D-aux__dr.png

    Harmonic Distortion:

    C5D-aux__thd.png

    Intermodulation Distortion:

    C5D-aux__imd.png

    Intermodulation Distortion (swept frequency):

    C5D-aux__imdsf.png

    -----------------------------------------------------------------------

    EDITED: added bass boost settings and RMAA info.
    EDITED: added more RMAA info.
     
    Last edited: Oct 18, 2017
    bartzky and HiFiChris like this.
  5. HiFiChris Contributor
    My fuzzy brain totally forgot about this thread's existence. Thanks, @hakuzen, I've added it to the overview in post #3. Just one little addition: if it's not too much extra work for you, seeing a graph of the bass boost implementation would be nice. :)
     
    Last edited: Oct 10, 2017
    hakuzen likes this.
  6. Raketen
    +1, C5D bass boost is really nice, would be cool to see DIY measurements. JDS own graphs show 14-16 dB max bass boost depending on gain. Their blog post on the C5D has a bunch of other measurements ( http://blog.jdslabs.com/?p=722/introducing-c5d-ampdac/ ), wish more manufacturers took a similar approach in marketing to explaining their products & engineering decisions etc...
     
    Last edited: Oct 10, 2017
    hakuzen and HiFiChris like this.
  7. castleofargh Contributor
    I've come to simply trust JDS when it comes to specs and measurements. they know what they're doing and have never been big on dishonest marketing game about numbers.
     
    HiFiChris, getclikinagas and hakuzen like this.
  8. hakuzen
    added the bass info. it doesn't match exactly the data in their blog.. maybe i did something wrong..
     
    HiFiChris and Raketen like this.
  9. HiFiChris Contributor
    I don't think that you did anything wrong. They probably changed the intensity and slope.
     
    hakuzen likes this.
  10. Samueru Sama
    Hey haku, what's the reference in those measurements (The 0dB output voltage of the device when it was measured)? So i can have an idea of the noise levels.
     
    hakuzen likes this.
  11. hakuzen
    Didn't expect less from my meticulous teacher.. You always make me go further and learn, maestro.
    Added more RMAA info about the headphones amplifier of JDS Labs C5D. Btw, in case of having to choose an unique voltage output to test, which one should be the reference?
    This extra info goes beyond the purpose of the thread, but now I got decent gear to measure, wanted to tease a bit.. hope that Chris forgives me..
     
    Samueru Sama likes this.
  12. Samueru Sama
    Either dBUu(0.775V) or dBV (1V) could be used, however smartphones may not be able to reach those voltages. I would always go for dBV and state the reference when that's not possible.

    Jds specs -105 dBu (4.3 uVrms) for the headphone amp and -103 dBu (5.4 uVrms) for the dac, Your measurements show 5.9 uVrms for the dac and 5.8 uVrms for the amp. Which is very nice and close to the official specs.

    If these measurements were unloaded, adding an small load (10K Ohm) to the c5d may make the noise figures even lower, the E-MU has a high input impedance and its susceptible to noise.
     
    hakuzen likes this.
  13. bartzky
    I personally use 0 dBu, because 0 dBV is the output limit of many small DAPs or smartphones or even not reached at all as @Samueru Sama already said.

    Here are my results for the JDS Labs C5D (DAC) at 0 dBu with no load and 30 ohm load, performed with a RME ADI-2 Pro:

    C5D.PNG

    So about 2.7 uVrms which is even better than JDS specs. The ADI's input noise is at about -114 to -115 dBu btw.
     
    Samueru Sama and hakuzen like this.
  14. castleofargh Contributor
    good catch, the emu 0404's line-in is specced at 1Mohm (add your favorite "your mama" joke). I always assumed ADC line inputs were around 10kohm for some reason because that's always what I've had on mine(real line-in, not crap mic-in from my laptop).
     
    hakuzen likes this.
  15. hakuzen
    gotcha, thank you all!
    i'll aim to dBu then (0.775Vrms) when possible (in case of DAPs and smartphones which use volume steps, guess nearest voltage to 0dBu will be enough, while stating it).
    shall use a load of 10KΩ (as "unloaded") if using E-MU 0404, to get closer to most ADCs results, and shall measure with 16-32Ω load as well.
     

Share This Page